(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Biodiversity Heritage Library | Children's Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Farm level analysis of irrigated crop production in areas with salinity and drainage problems"



^'i^i\:t I~jf 



Prepared Under Contract for the 

Federal-State 

San Joaquin Valley 
Drainage Program 



FARM-LEVEL ANALYSIS OF 

IRRIGATED CROP PRODUCTION IN 

AREAS WITH SALINITY AND 

DRAINAGE PROBLEMS 



March 1 988 



This report presents the results of a study conducted for 
the Federal -State Interagency San Joaquin Valley Drainage 
Program. The purpose of the report is to provide the Drainage 
Program agencies with information for consideration in 
developing alternatives for agricultural drainage water 
management. Publication of any findings or recommendations in 
this report should not be construed as representing the 
concurrence of the Program agencies. Also, mention of trade 
names or commercial products does not constitute agency 
endorsement or recommendation. 



The San Joaquin Valley Drainage Program was established in 
mid-1984 as a cooperative effort of the U.S. Bureau of Reclamation, 
U.S. Fish and Wildlife Service, U.S. Geological Survey, California 
Department of Fish and Game, and California Department of Water 
Resources. The purposes of the Program are to investigate the 
problems associated with the drainage of irrigated agricultural lands 
in the San Joaquin Valley and to formulate, evaluate, and recommend 
alternatives for the immediate and long-term management of those 
problems. Consistent with these purposes. Program objectives address 
the following key areas: (1) Public health, (2) surface- and ground- 
water resources, (3) agricultural productivity, and (4) fish and 
wildlife resources. 

Inquiries concerning the San Joaquin Valley Drainage Program may 
be directed to: 

San Joaquin Valley Drainage Program 
2800 Cottage Way, Room W-2143 
Sacramento, California 95825-1898 



Areas 



Farm Level Analysis of 
Irrigated Crop Production in 
with Salinity and Drainage Problems 



Prepared for the 

San Joaquin Valley Drainage Program 
2800 Cottage Way, Room W-2143 
Sacramento, California 95825-1898 



Under 



U.S. Bureau of 
Cooperative Agreement 



Reel amat i on 

No. 7-FC-20-04990 



By 



Denni s Wi chel ns 
Department of Resource Economics 
University of Rhode Island 



Dani el Nel son 

San Luis Water District 

Los Banos , Cal i forni a 



Thomas Weaver 
Department of Resource Economics 
University of Rhode Island 



January 1988 



Farm-level Analysis of Irrigated Crop Production 
In Areas with Salinity and Drainage Problems 



Table of Contents 

Section ll21 

1. Farm- 1 eve! Economic Impacts of Salinity 2 

1.1 Yield and acreage data 4 

1.2 Yields and revenues, by period 6 

1.3 Summary 1*^ 

2. Crop-specific Production Functions 12 

2.1 A multicrop production model 13 

2.2 Estimated production functions 15 

2.3 Estimated water application rates 17 

2.4 Summary 1° 

3. Grower Interviews 19 



3.1 Land qua I ity 

3.2 Crop rotations 

3.3 Pre-i rrigation 

3.4 Drainage 

3.5 Grower experience 



20 
21 
22 
23 
23 



3.6 Summary 24 

4. Soils Data Collection and Analysis 26 



4.1 Methodology 27 

4.2 Soil salinity, sodum adsorption ratio, and boron 29 

4.3 Selenium 31 

4.4 Electromagnetic meter measurements 33 

4.5 Summary 34 

5. Current Efforts and Future Work 36 

5.1 Activities 36 

5.2 Discussion 38 

5.3 Summary ^1 

6. References ^2 

7. Tables 43 

8. Appendix 1: Selected Grower Comments 57 

9. Appendix 2: (Quality Assurance Information 65 

10. Figures 69 



Farm-level Analysis of Irrigated Crop Production 
In Areas with Salinity and Drainage Problems 



Farmers affected by so i I salinity and high water tables can alter 
cropping patterns and water use to maintain productivity in the short 
term. Leaching water will remove salts from the root zone and 
subsurface drainage systems can be installed to manage the soil water 
level, in fields with perched water tables. These practices are 
effective when the salt content of irrigation water is not excessive and 
when adequate drainage and disposal facilities exist. In recent years, 
the environmental and technical costs of developing drainage outlets 
and/or treatment facilities have received increased attention. Some of 
these costs have resulted in restrictions on drainage outflows in some 
locations. 

The purpose of this research is to examine farm-level responses to 
worsening salinity and drainage conditions and to provide information on 
the likely responsiveness of growers to policy alternatives which may be 
proposed in order to manage the quantity and quality of drainage water 
generated on a local or regional basis. Results will be useful in 
formulating recommendations for growers faced with changes in drainage 
outflow constraints and in evaluating regional models of agricultural 
production and hydrologic balances. 

This report describes progress made to date in four areas of 
research which will contribute to the above objectives: 



1. Examination of farm-level economic impacts of salinity and high 
water tables. 

2. Estimation of crop-specific production functions for crops 
typically grown in drainage problem areas of the San Joaquin 
Va I I ey . 



3. Interviews with growers who have produced crops in areas with 
and without salinity and drainage problems. 

4. Collection and analysis of soil and water quality data in a 
drainage problem area. 

A fifth section of the report describes how the current efforts and 

results will be useful in future work to complete the goals of the 

project. 

1. Farm- 1 eve I Economic Impacts of Salinity 

Yields of salt-sensitive crops will decline if soil and water 
salinity cannot be managed effectively. Farmers will shift to growing 
salt tolerant crops when it is no longer profitable to produce the 
sensitive ones. The ultimate economic effects of salinity can therefore 
be described by examining changes in yields and cropping patterns, over 
time. Observed changes in these, however, may also be caused by market 
factors such as changes in relative prices. Salinity effects need to be 
separated from market effects. 

Some of the costs associated with crop yield losses due to salinity 
and high water tables are examined in this study. Farm-level, time- 
series data on yields and cropping patterns in an irrigation district on 
the west side of California's San Joaquin Valley are analyzed. The 
district is in an area known for its problems with salinity and high 
water tables. Changes in crop yields and the average value of land 
productivity, over time, are compared with trends throughout a larger 
production area. The purpose is to identify yield effects that may be 
due to salts and high water tables in the smaller district. 
Background 

The 10,000-acre Broadview Water District is located in the northwest 
corner of Fresno County. Surface water is the major source for 



irrigation in Broadview and has generally been available in adequate 
amounts since 1957. In contrast, a drainage outlet has not always been 
available and the district had to recycle all of its mingled surface 
(tailwater) and subsurface drainage water through 1982. This resulted 
in the application of high-salt irrigation water and the accumulation of 
salts in district soils. 

Broadview's fresh water supply is delivered from the Sacramento 
River Delta, via the federal Del ta-Mendota Canal. The average salinity 
level of the fresh water supply ranges from 200 to 400 parts per million 
of total dissolved solids. These values are typical of good quality 
canal water in the San Joaquin Valley. 

Water delivered to Broadview growers is a combination of fresh water 
and mingled surface and subsurface drainage waters collected in the main 
district drain. The district had to recycle all of its high-salt drain 
water for many years, since a disposal outlet was not available until 
1983. The ratio of drain water to fresh water in quantities actually 
delivered to growers increased, over time, from near zero in the early 
1960's to about 50% in the early 1980's. 

District personnel began collecting soil and water salinity 
information in 1980. The average level of salinity in the Broadview 
drain ranged from 2,700 to 2,960 parts per million, total dissolved 
solids, from 1980 through 1982. Several readings indicate salinity 
levels greater than 5,000 parts per million. Fresh canal water 
salinity averaged from between 300 to 350 parts per million, during this 
period. The blended drainage and fresh water delivered to growers 
occasionally exceeded 5,000 ppm, but averaged from between 1,800 to 
2,150 parts per million. These measurements imply that salts had 



accumulated in district soils and contributed to significant increases 
in the salinity level of water applied to crops. 

Yield and Acreage Data 

Farm- level data on crop acreages and yields in Broadview were 
collected from annual reports submitted to the district. Throughout the 
data collection process, district growers suggested that cropping 
patterns had shifted significantly during the 1970's. The consensus was 
that high-value crops such as tomatoes, melons, and alfalfa seed had 
been replaced with lower value, salt-tolerant cotton and sugarbeets as 
soil salinity problems increased. 

Annual crop production and yield data for Fresno County were 
obtained from the agricultural commissioner's reports. These data 
provide information on the total acreage and average yields of crops 
grown in the county. Unlike Broadview, most other areas of the county 
are not adversely affected by salinity or high water tables. Since 
growers throughout the county face similar technological and market 
conditions, significant differences between the county and the district 
in yields and cropping patterns may be attributable to deteriorating 
soil and water conditions in Broadview. 

Cotton acreage increased over time in both Broadview and Fresno 
County as whole. Differing trends are apparent for tomatoes and alfalfa 
seed (Figure 1). Acreages of these crops countywide either remained 
constant (tomatoes) or increased (alfalfa seed) after the early 1970's, 
but they declined in Broadview. Tomato acreage in Broadview was 
greatest in 1973, declining to zero in 1982. Alfalfa seed acreage fell 
from nearly 4,000 in 1970 to 560 in 1982. Acres in sugarbeets have 
remained constant in Fresno County but have increased in the water 
district. 



These differing trends for tomatoes and alfalfa seed motivate 
examination of crop yields in the two areas to determine if these were 
also deverging, over time. Cotton yields per acre have generally been 
higher in Broadview than in Fresno County. Yields of tomatoes in 
Broadview were consistently higher than the county average through 1973. 
District tomato yields were highest in the middle to late 1960's and 
declined in the 1970's. County yields were at or above the Broadview 
level from 1974 through 1982. 

Yields of alfalfa seed increased in the early 1960's in Broadview, 
before falling at the end of the decade. Despite this decline, yields 
were at or above those in Fresno County for the 11 years including 1971 
through 1981. 

One important reason for the reduced alfalfa seed yields and acreage 
in Broadview in 1969 and the early 1970's was the decline of alkali bee 
populations. Broadview growers report that these bees are much 
preferred to normal honey bees for pollination, because (1) alkali bees 
are less distracted by blossoms of other crops in neighboring fields, 
and (2) they are better adapted to and more persistent in the difficult 
task of distributing alfalfa seed pollen. According to growers, bees 
are a key input in growing alalfa seed, and difficulties in obtaining 
good pollination can significantly affect crop yields. 

Barley and wheat yields have generally been higher in Broadview than 
in Fresno County, with exceptions in some years. Sugarbeet yields were 
higher in Broadview through 1974, while county averages were greater in 
1976, 1979, and 1980. 

Examination of crop yields from a time-trend perspective suggests 
important differences that evolved between 1962 and 1982. Cotton and 



barley yields remained relatively constant in both locations throughout 
these years. Tomato yields increased in Fresno County while declining 
in Broadview. Alfalfa seed yields remained constant in the county while 
declining in the water district. Wheat yields display a positive time 
trend in each region. Sugarbeet yields are described by a positive 
trend in Fresno County, while a flat time line is suggested for 
sugarbeets in Broadview. 

Yields and Revenues, By Period 

Acreage and yield trends in Broadview and Fresno County suggest 
important differences for some crops. The crops examined here represent 
varying degrees of salt sensitivity. Barley, wheat, cotton, and 
sugarbeets are generally regarded as salt tolerant crops, while tomatoes 
and alfalfa seed are relatively sensitive (Maas and Hoffman). Site- 
specific soil and water table conditions, irrigation water quality, and 
management practices can influence the impact which salts have on 
yields and the longevity of crop production. 

Broadview cropping patterns indicate increasing acreages of cotton 
over time as the importance of other crops has diminished. These 
changes may be caused by both technological and market factors, in 
addition to the effects of salinity and high water tables. A closer 
review of acreages and yields, by time period, suggests which crops may 
be responding to so i I salinity and high water tables. 

Information on the history of Broadview's re-circulating program 
motivates examination of crop yields in four separate time periods. 
These are chosen to correlate with distinct changes in cropping 
patterns, subsurface drainage installations, and drainage outlet 
availability in the Broadview Water District. The periods are 1962 



through 1970, 1971 through 1976, 1977 through 1982, and 1983 through 
1986. 

The first period includes the years prior to installation of 
subsurface drains in the district. Barley, cotton, and melons were the 
major crops grown initially. Alfalfa seed acreage increased during this 
period, while melons and barley gradually declined. Tomatoes were 
introduced in the late 1960's, but acreage of this crop declined 
briefly, prior to the installation of subsurface drains. 

The second period is one in which leaching fractions (the portion of 
water applied in excess of plant requirements) increased, as subsurface 
drains were installed and the flushing of salts became feasible. Tomato 
acreage reached a peak in 1973 and remained high through 1975. Alfalfa 
seed acreage declined throughout the period, while cotton plantings 
increased moderately. The third period, 1977 through 1982, was 
dominated by cotton production, as tomato and alfalfa seed acreage 
declined to near-zero levels. 

Broadview's drainage outlet was constructed in 1982 and the first 
drain water releases occurred in January of 1983. The fourth time 
period includes the first four years of crop production following 
installation of a drainage outlet. During this period, the quality of 
irrigation water delivered to growers improved markedly, as smaller 
amounts of drain water were blended into the delivery canal. The 
average salinity of delivered water fell from 1,809 parts per million, 
total dissolved solids, in 1982 to 592 parts per million in 1983 and 
remained in the 400 part per million range for the next three years. 
This occurred as the portion of mingled drainage water in total 
deliveries declined from 41 percent in 1982 to 18 percent in 1983. 



8 

The average salinity of mingled tailwater and subsurface drain water 
fell from 2,713 parts per million in 1982 to 2,284 ppm in 1983. The 
average rose again in 1984, but declined in both 1985 and 1986. This 
downward trend in drain water salinity may indicate that soil salt 
concentrations were declining, as accumulated salts were finally being 
removed from district soils. 

Broadview tomato acreage increased from zero in 1982 to 750 acres in 
both 1985 and 1986. Melon plantings increased from 455 acres in 1982 to 
790 acres in 1985 and 670 acres in 1986. Acreage in alfalfa seed grew 
from 560 in 1982 to 630 acres in 1985 and 705 acres in 1986. 

Average Broadview yields of selected crops during the four time 
periods are compared with those for Fresno County in Table 1. Cotton, 
alfalfa seed, barley, and wheat yields follow similar time trends in 
each area, with the Broadview average yields being higher than those in 
the county. Average yields of tomatoes and sugarbeets follow different 
trends. The average tomato yield in Broadview declined in both the 
second and third periods, while these averages increased in Fresno 
County. The average sugarbeet yield declined in the third period in 
Broadview, while rising in the remainder of the county. Average yields 
of all crops increased between the third and fourth periods in both 
locations. 

Observed changes in average crop yields represent one of the factors 
which determine the relative profitability of crop alternatives. Other 
factors include changes in relative prices. Prices reported for crops 
in Fresno County, over time, are generally similar to those received by 
Broadview growers. 



Average per-acre revenues for the selected crops are calculated by 
multiplying average yields by the average prices received (Table 2). 
Cotton and wheat revenues are consistently higher in the district than 
in Fresno County. Broadview nominal returns from alfalfa seed 
production are at or above the county average in all periods. Tomato, 
barley, and sugarbeet returns fall below the county average in the third 
period. Revenues for all crops are greater in Broadview than in Fresno 
County, following improvements in the drainage situation. 

The average per-acre cotton revenue increased by 57 percent in 
Broadview between the second and third periods, while tomato returns 
declined by 8 percent. These trends may indicate a change in the 
relative profitability of growing tomatoes and cotton in Broadview 
during these periods. Alfalfa seed and sugarbeet revenues were also 
rising at a slower rate than cotton returns between periods two and 
three. 

One estimate of the economic value of a drainage outlet in Broadview 
is obtained by examining the yield improvements between the third and 
fourth periods. These increases are adjusted downward to reflect 
percentage improvements in county average yields that also occurred 
during those years. This is done to allow for improvements in 
technology, prevailing weather conditions, and/or reductions in pest 
problems. The net yield increase in Broadview is then attributed to 
improved soi I and water conditions. 

Percentage yield increases are greater in Broadview than in Fresno 
County for all crops except cotton (Table 3). In Broadview, the net 
yield increase, or portion of the 15.5-ton increase in tomato yields 
attributed to drainage improvements, is 13.5 tons (Table 4). The value 



10 

of this increase, at the average nominal price received during 1983 
through 1986, is $698 per acre. Similar values range from $37 per acre 
for wheat to $109 for alfalfa seed. 

Summary 

Compelling evidence of crop responsiveness to improvements in 
drainage conditions is provided by increases in average crop yields for 
the four-year period following installation of a drainage outlet in the 
Broadview Water District. Average yields of the six crops examined here 
exceeded county averages during these years. Yields of salt-sensitive 
tomatoes and alfalfa seed responded by 80 percent and 56 percent, 
respectively. Average yields of salt-tolerant barley, wheat, and 
sugarbeets have also improved in Broadview, at a rate exceeding Fresno 
County yield increases, during the most recent period. This suggests 
that even crops which are considered to be tolerant of soil salinity may 
be adversely affected by continuous recycling of all drain water, over a 
prolonged period of time. Yields of these crops have responded 
positively to improvements in the quality of irrigation water and 
reductions in soil salinity, made possible by improving the drainage 
situation in Broadview. 

The value of obtaining the drainage outlet and improving soil and 
water conditions is estimated for selected crops. These data, while 
pertaining specifically to salinity conditions and yield improvements in 
the Broadview Water District, provide an indication of the costs 
associated with yield reductions due to salinity and high water tables. 
An additional indication is that the effectiveness of leaching programs 
may be limited in an area which does not have a drainage outlet. Cotton 
and wheat yields remained constant or increased after subsurface drains 



11 

were installed in Broadview, but yields of tomatoes, alfalfa seed, and 
sugarbeets declined prior to installation of the drainage outlet. 

Detailed analysis of the production relationships underlying the 
observed changes in Broadview crop yields, over time, is described in 
the following section of the report. 



12 

2. Crop-specific Production Functions 

Farm-level crop production functions describe the responsiveness of 
crop yields to changes in the level of inputs and other factors. These 
are often estimated using farm-level data on yields, input use, and soil 
and water quality factors, when these data are available. Crop-specific 
production functions describing yields as a function of the quantity and 
quality of applied water and soil quality are useful in predicting crop 
responsiveness to changes in drainage situations. Production functions 
are also an integral part of any behavioral mode! of grower response to 
salinity and drainage policy alternatives. 

The relationship between inputs applied in production of crops and 
the yields produced may change, over time, in response to deteriorating 
soil and water conditions. Estimation of this relationship in a way 
that Includes the effects of soil and water quality is desirable. 

In the absence of detailed data on soil and water quality, 
adjustments must be made in the estimation procedure to account for the 
effects of these variables. One useful approach is to gather 
information from growers which describes the effect of changing soil and 
water quality on crop production practices and yields. Another approach 
is to examine historical data to determine when significant changes in 
quality variables may have occurred. Shift variables pertaining to 
identifi-ed time periods can then be included in an estimation procedure. 

Distinct changes in salinity and drainage conditions in the 
Broadview Water District, during four time periods, are described in the 
first section of this report. Crop-specific production functions are 
estimated for these time periods, allowing shifts in the relationships 
to correspond to changes in soil and water conditions. Estimated shifts 



13 

in the production functions are used to describe the empirical effects 
of salinity and high water tables on crop yields. A multicrop 
production framework which is appropriate for use with the data 
available from Broadview is used for this analysis. 

A Multicrop Production Model 

The data collected from Broadview Water District include crop 
acreages, yields, revenues, and the total amount of water applied by 
each grower. Crop-specif ic water application rates were not recorded 
until 1985. Hence, an empirical framework allowing limited observations 
on variable input use is required. The multicrop production model 
presented by Just, Zilberman, and Hochman is appropriate for this 
empirical situation, since certain variable input use rates are not 
observed. This framework is used here to estimate production function 
parameters and the unobserved crop-specific irrigation rates. 

Farmers are assumed to be maximizing expected net revenues, subject 
to production constraints: 

(1) max E(t) = E(p y - W2X2) 

Y/ '^> ^2 

subject to: y = f (X) 

Xi, = X 

where: y = (y, , • . ., Yu) , a vector of 
farm outputs; 

X = (x- , x_) , a vector of land and 
water inputs; 

X = [x.,], a (2 by K) matrix of 
i nput a I locations; 

L = (1, 1, . . ., 1)^; 

w^ = price of irrigation water; 
r = net revenue; 



14 



Cobb-Douglas production functions for the case of two inputs, land 
and water, are specified as: 

(2) y-.k = ^lk ^2k ^ 

where: y., = output of the kth crop for farmer i; 
X.-, = acreage planted to the kth crop; 
x.„, = irrigation water applied to the crop; 

a-, = production elasticity, j = 1, 2; 

J "^ 

€■, - a random error term for the kth 
production function, N(0,v,); 

e = the natural exponent. 



Output price is a random variable and is treated as in Just, 
Zilberman, and Hochman. Marginal value product relationships are 
substituted for the unobserved irrigation water applications, according 
to first-order conditions: 



'^^k ^ik 
(3) x.^;^ = ^' , for k = 1 K; 

*2 



where r., = revenue received by the ith grower for the 
kth crop. 



These expressions are used to form the following system: 

K r.^ 

(4) ^.^^i .-L. , g.^x 

k=l ^2 



r . , 
(5) In y;^ = a^^ In x.^^ * a^k '" " * '^2k '" "2k * 4 * ^ i k^ 

In equations (4) and (5), marginal value product relationships are 
substituted for the unobserved water allocations among crops. Equation 



15 

(4) is a variable input summation equation, reflecting the fact that 
water applied to a I I crops must equal the total amount purchased. An 
additive error term, e.„ , is included in this equation since actual 
data will rarely fit this relationship exactly. 

Equation (5) represents the crop-specific production functions. 
These equations are not directly estimable due to the unobservabi li ty of 
expected revenue. This is replaced with actual revenue and a set of 
appropriate error terms, as described in Just, Zilberman, and Hochman. 
An intercept term, p., is included in Equation 5, as part of the 
necessary modifications. 

Full information estimation of this system is achieved using 
nonlinear three-stage least squares. 

Estimated Production Functions 

Broadview farm-level data on crop acreages, yields, revenues, and 
aggregate water use and expenditure for the period 1962 to 1986 are 
analyzed. Production functions are estimated for the major crops grown 
in the district during each of the four periods described above. 
Irrigation water and land are the two inputs considered. Constant 
returns to scale are imposed on the estimated relationships. 

Estimated regression parameters for five crops grown in all four 
time periods are presented in Table 5. Plots of the estimated functions 
appear in Figures 2a through 2e. 

The barley function does not shift significantly between any of the 
four time periods, as indicated by the shift F-statistics. The 
estimated water elasticity for 1983 to 1986, 0.324, is highest of the 
periods examined. Barley is grown least frequently during these years, 
resulting in a low number of observations for this crop. 



16 

The estimated cotton production function shifts significantly 
outward between the third and fourth periods. Both the constant term 
and the water elasticity are higher in the final period than in the 
previous one. 

The estimated production function for sugarbeets shifts outward 
between the first and second periods, while remaining movements are not 
statistically significant. The number of observations used to estimate 
the sugarbeet function is smallest of the five crops examined, in the 
first three periods. 

The tomato and alfalfa seed production functions display similar 
shifts, over time. Both curves shift inward from the first period 
through the third, before shifting outward in the final years. The 
initial shift of the alfalfa seed function is not statistically 
significant, but all other shifts of these functions do provide better 
explanation of yield variability than does a single function for any two 
of the time periods. 

Estimated production functions for six additional crops grown 
during selected time periods are presented in Table 6. The outward 
shift in the wheat function between the third and fourth periods, though 
not statistically significant, appears to suggest greater production 
potential for this crop in the years following installation of the 
district's drainage outlet (Figure 2f ) . 

In general, the estimated production functions are consistent with 
changes in yields which have occurred in the water district, over time. 
Salt-sensitive tomatoes and alfalfa seed production parameters are 
greatest in the final period, following installation of the drainage 
outlet. Salt-tolerant cotton and wheat functions are also shifted 



17 

outward in this period. Apparent shifts in the barley production 
function are not statistically significant, but the responsiveness of 
yields to applied water appears to be greatest in the final period. 
Similarly, recent shifts in the estimated sugarbeet function are not 
significant, but the number of observations for this crop is often 
lowest of those examined. 

Estimated Water Application Rates 

The amount of water applied per acre, by crop, was not reported in 
the data. Yet this information is important in evaluating the 
responsiveness of growers to drainage management alternatives. The 
estimated water elasticities are used to estimate the unobserved, crop- 
specific water application rates, using the marginal value product 
rel ati onsh i p : 



(9) ^2k = "IV 



-"ik 



*2 



where: '"•■')> = estimated irrigation water applied by the 
ith grower on the kth crop; 

Ort. = estimated water elasticity for the kth crop; 

r.. = revenue received by the ith grower for sale 
of output from the kth crop; 

Estimated district average water applications for selected crops 
are compared with actual district averages available for 1985 and 1986 
in Table 7. Estimated water application rates for cotton, sugarbeets, 
and wheat compare favorably with the actual amounts applied in 1985 and 
1986. Similarly, the estimated 3.77 acre feet per acre applied to 
"other" crops is near the actual 3.75 acre feet applied on corn in 1985 
and the 4.18 acre feet applied on beans in 1986. Estimated barley and 



18 

alfalfa seed applications are higher than those actually recorded in the 
two most recent years. The tomato estimate is lower than actual 
irrigation amounts. The disparity between estimated and actual 
irrigation rates for barley and tomatoes suggests that additional 
information would be helpful in estimating production functions for 
these crops. 

Summary 

Time-series farm-level data on crop yields, revenues, and aggregate 
water applications are used to estimate crop-specific production 
functions for major crops grown in the Broadview Water District. 
Information from growers and water district personnel is used to 
describe four time periods during which distinct changes in salinity and 
drainage conditions have occurred. Significant shifts in the production 
relationship are identified for both salt-tolerant and salt-sensitive 
crops. The estimated functions are consistent with observed changes in 
average yields in the district, over time. 

Unobserved water applications, by crop, are estimated and compared 
with actual irrigation rates for the two most recent years. Estimates 
compare favorably with actual applications for three of the six crops 
examined. Additional field-specific and crop-specific data will be 
useful in obtaining more accurate estimates of irrigation and production 
relationships. 

Two efforts designed to collect information which will enhance the 
appropriateness and accuracy of the empirical analysis are described in 
the following sections of the report. These are interviews with 
Broadview growers and collection of field-specific soil data. 



19 

3. Grower Interviews 

Eight of the fourteen growers who operate farms in the Broadview 
Water District were interviewed in August, 1987, using an ethnographic 
interview format. These eight growers farm about 6,930 acres, or 
approximately 73 percent of the total irrigated acreage in the district. 
The ethnographic interview technique is designed to elicit growers' 
resource taxonomies, in their own words. The effects which individual 
elements in any taxonomy have on farm-level allocation decisions are 
then assessed. Within this methodology, it is assumed that (1) growers 
allocate resources based on their individual perceptions of the 
resources they command, (2) that these perceptions are dynamic and 
change in response to such factors as increased experience, 
technological change, new information, and government regulation and 
policy, and (3) that these perceptions identify the relevant variables 
in their assumed production relationships and the resource constraints 
that determine farm-level resource allocation. 

The objectives of this research are the following: 



1. To gain insights into the nature of on-farm water management in 
the district. This information will be useful in identifying 
water management policy options and for predicting grower 
response to these alternatives. 

2. To compare information from growers regarding land 
characteristics to field-level data obtained from soil sampling 
and laboratory analysis. This will assist in improving the 
efficiency of collecting information on drainage problems and 
the consequences of water management policies. 

3. To provide guidance for any further surveys of a wider 
geographical area. 



Information provided by growers is summarized and discussed below in the 
context of farm management considerations. These include land quality, 



20 

crop rotations, pre-i rrigatlon, drainage system performance, and grower 
experience. 

Land Qua I i ty 

Information detailing some of the characteristics of quarter 
sections operated by growers interviewed during the survey reveals 
several important points. It is apparent that individual quarter 
sections are the primary management units for most of the farmers in the 
district. Growers identify a set of characteristics describing each 
quarter section of land (or smaller units, in the case of some irregular 
holdings on the border of the district). These determine the manner in 
which they will manage the land in order to grow a particular crop. In 
those few instances where quarter sections are divided into smaller 
units, land characteristics, such as large sand lenses or salt problems, 
seem to be the basis for the division. 

The soil categories — heavy clay, light clay, sandy and so on — are a 
subjective judgement on the part of the grower and are recorded here in 
the growers' terminology. Soil taxonomies are of interest to the extent 
that soil differences affect management decisions. It is apparent from 
growers' comments that soil type, as defined by them, is important in 
determining water application strategies and in controlling salt build- 
up. Heaviness of the soil, for example, affects the timing of pre- 
irrigation in cotton. Heavy soils with poor internal drainage must be 
pre-i rrigated during the fall or winter months if the fields are to be 
dry enough in the spring to permit timely planting of the crop. 

Based on our consideration of the survey data to date, it appears 
that each quarter section, or in some cases a smaller field, is viewed 
by the grower as having crop production possibilities based on its water 



21 

holding characteristics and its salt content at any point in time. Salt 
content is a function of the quality and quantity of irrigation water 
that has been applied in the past and the drainage characteristics of 
the fields. These, in turn, are considered to be determined by the 
basic soil, the field's location, slope and water table (all of which 
influence drainage), and are subject to control by the addition of 
gypsum as a soil amendment and the installation of drainage tiles. The 
effectiveness of the drainage tile is considered to be determined by the 
depth and density of the tiles and the precision with which the system 
was i nsta I led . 

Crop Rotations 

The particular field circumstances, along with the usual factors 
such as government programs (eg. acreage set aside), the availability of 
contracts with processors, and prices determine the crops which the 
grower chooses to grow. The capacity to control water in order to 
achieve the desired field conditions deemed favorable for plant growth, 
is considered by growers to be more important than all other factors, in 
achieving high yields and net returns. 

There is widespread concern among growers regarding soil quality. 
All of the growers who were interviewed expressed some concern over 
managing their fields in a way that maintains desirable soil conditions. 
Some growers keep wheat in their crop rotation, even though they feel it 
is not the crop which will maximize profits, at current prices. The 
value of wheat in a crop rotation is that the crop residue contributes 
to soil fertility and conditioning, and thus improves the yields of 
subsequent crops. In other words, growing wheat is a cost of the 
production of other crops. Chicken manure and gypsum are seen as soil 



22 

improvement materials. A number of growers have regular programs for 
incorporating these additives. 

It is clear that salts play an important role in determining the 
growers' crop rotations. Tomatoes are known as a particularly salt- 
sensitive crop, although the availability of contracts with canneries 
also acts to restrict tomato acreage. Sugarbeets, on the other hand, 
perform well on salty ground but have a high crop water requirement (up 
to five acre feet per acre, to mature a crop). 
Pre-i rr igation 

Pre-i rr igation is important in controlling salinity and in providing 
water for establishing the crop. The amount of water applied and the 
timing of pre- i rr igati on depend on the field circumstances and the crop 
to be grown. For example, wheat is generally planted in late fall/early 
winter with no pre- i rr i gati on . Similarly, alfalfa seed, and sometimes 
tomatoes, are brought up on rainfall. 

There is considerable variation in the timing of pre-i rr igation for 
cotton. Some growers pre-irrigate in late fall/early winter. Two 
rationales are given for this practice. One grower, whose fields are 
near the end of the district's distribution system, pre-i rrigates early 
because competition for water is minimal at this time of year and also 
because the quality of the water is better than during later months. 
This is because the percentage of re-circulated subsurface water in the 
system is low at this time of year. Other growers pre-irrigate early 
because their soils are such that if they wait until later in the 
season, the fields will not be dry enough to move in the equipment for 
spring planting. This is typically an important consideration on the 
heavier clay soils and those with a salt problem, where it is difficult 



23 

to get water to move into the soil profile. Delaying pre-i rrigation to 
the early spring is not an option available to growers faced with these 
ci rcumstances. 

In general, growers recognize the potential utility of sprinkler 
systems for the first irrigation on cotton in the spring, and for 
i rrigati ng-up other spring-planted crops. Growers are not in agreement, 
however, that the water savings and the benefits to the crop will 
justify the necessary investment. Growers see sprinkling as a good way 
to avoid salt accumulation in the seed beds when furrow irrigation is 
used for pre-i rrigati ng. With sprinklers, the salts are dispersed 
during the first irrigation, to the benefit of the plant. For 
subsequent irrigations on crops such as cotton, furrows are considered 
the appropriate technique. 

Dra i nage 

One important constraint to crop production and to water management 
and conservation, according to growers, is the presence and adequate 
performance of tile drainage systems. Several growers are not satisfied 
with the effectiveness of their tile drains and blame faulty design 
and/or installation. Some growers suggest that improvements in drainage 
conditions would result in reduced water applied during pre-i rrigation . 
Overall, water management , including both field application and 
drainage, appears to have an importance to the growers which is greater 
than that of any other production input. 

Grower Experience 

One general impression of agricultural production in the Broadview 
Water District is that each of the farmers has developed, over time, a 
production system which is sensitive to the constraints encountered in 



24 

their fields, and which also includes the grower's preferences for crops 
with which they have been successful. For example, there have been only 
three farmers in the district over the last several decades who have 
been alfalfa seed growers. Currently there are two, and each speaks to 
the difficulties of managing the crop for high production levels. They 
suggest that it requires a specialized knowledge. 

Other examples of growers' preferences include statements such as, 
"I don't grow tomatoes because I don't want to go broke," or "I can do 
good at small grains, but there is no money in them," or "I like to grow 
melons," or "Really, we are primarily rice growers," and so on. In 
other words, the enterprise combinations that one finds on a given farm 
are the result of the grower's total experience with working that farm. 
This experience includes successes, prejudices, and preferences. A set 
of selected grower comments describing many aspects of grower 
experiences are presented in Appendix 1. 

Summary 

Of special interest to this study are the water conservation 
techniques that have been adopted by growers in the Broadview Water 
District. To date, our analysis of the data does not allow 
identification of these different practices with the cultivation of 
different crops under different soil and salt conditions. However, the 
hypothesis at this time is that there is a relationship between water 
management practices, crops, and soils. The following observations are 
relevant to the water management issue: 

1. Almost all of the Broadview growers use quarter-mile runs for 
pre-i rr igation. 

2. Growers have been experimenting with different irrigation 
strategies on cotton, including irrigating in alternate furrows 



25 



and changing the time-length of irrigation sets as a means of 
conserving water. 

3. Sprinkler irrigation is viewed as a way of dealing with salt 
accumulations in the seed rows and of conserving water. 
Growers have different opinions as to the economics of 
sprinkler irrigation on the different crops and soil situations 
in the district. 

4. Several growers have inadequate drainage systems. In their 
opinion, this results in increased amounts of water used for 
pre-i rr igation. 

5. Neither gated pipe nor surge irrigation has taken hold in the 
Broadview Water District. One grower, interviewed from a 
neighboring district, is using "manual surge" to reduce 
irrigation water applications. He feels this technique is 
effective in reducing both tailwater and subsurface drainage. 

6. The strategy for pre-i rrigati ng cotton is definitely a function 
of the soil and salt conditions encountered by individual 
growers. 

Further analysis of the information gathered from grower interviews 
will include a comparison of measured physical characteristics with 
problem areas identified by the growers. Also, farm-level data will be 
used to identify appropriate variables and resource constraints for use 
in analysis of production relationships. 

A description of the soils data collection procedure and some 
preliminary analysis are presented in the following section. 



26 

4. Soils Data Collection and Analysis 

The agricultural drainage problem in the San Joaquin Valley and 
elsewhere arises from the need to leach accumulated salts from the soil 
profile. During the leaching and drainage process, these salts and 
other constituents are moved through the soil and into receiving pipes 
or ditches. Soil quality takes on a dual role in farm-level drainage 
management, as a result of environmental concerns regarding drain water 
qua I i ty . 

The level of nutrients, salts, and metals in agricultural soils 
affects plant growth and productivity. Farmers strive to maintain 
optimal levels of these soil constituents, in order to achieve 
profitable yields. The level of various materials in soils may also 
determine the quality of drainage water produced from a farm operation. 
There may also be an empirical relationship between the amount of 
leaching water applied to the soil profile and the quantity and quality 
of subsurface drainage water generated. 

Efforts by growers to modify the quantity or quality of subsurface 
drainage water leaving the farm will likely result in changes in soil 
quality. These changes may affect crop yields and the quality of 
drainage water produced in the future. Accurate description of 
relationships between soil quality, yields, and subsurface drainage 
water quality is necessary to fully describe potential farm-level 
responses to drainage management policies. 

Soils data are collected in this study for two purposes: 1) to 
enhance the production function analysis presented in the second part of 
this report, and 2) to examine empirical relationships between farm- 
level irrigation strategies and the quantity and quality of drainage 



27 



water produced. These efforts will improve the pertinence and validity 
of any recommendations which arise from the study. 
Methodology 

Soil samples were collected from agricultural fields in the 
Broadview Water District during the summer of 1987. Most of the fields 
are 160-acre quarter-sections of land and are planted to a single crop. 
Some fields are smaller than 160 acres and others are planted to two 
crops, or to one crop and a fallow area. In the 160-acre fields, five 
soil samples are collected along each of four east-west transects. 
These twenty samples include material from the top three feet of the 
soil profile. One additional sample is taken from three to six feet 
deep, at a location near the corner of the field. 

Soil salinity is measured during the data collection effort with an 
electromagnetic conductivity meter (EM-38) . This hand-held device is 
carried throughout the fields and readings are taken at each of the 
twenty sites. Both horizontal and vertical readings are taken, in order 
to characterize soil salinity from a depth of zero to 6 feet. An 
additional measurement is taken at the site where the 3 to 6 foot soil 
sample is gathered. 

The electromagnetic meter has been shown to generate reliable soil 
salinity measurements when used in conjunction with laboratory analysis 
of selected soil samples (Corwin and Rhoades, Rhoades, Rhoades and 
Corwin). The meter may become a useful tool for growers or water 
district personnel interested in monitoring soil salinities under 
different irrigation and drainage strategies. 

Four of the individual soil samples are sent directly to the 
laboratory for chemical analysis. These include samples from locations 



28 

which have the two highest EM readings and the lowest EM reading. The 
sample tal<en from the 3 to 6 foot depth is also sent directly to the 
lab. Five additional samples are prepared by combining soil from some 
of the individual sites, to form composite samples. Four of these are 
prepared by combining soil from the five samples taken along each 
transect, to form a single sample representing the transect. The final 
composite sample is made by combining soil from each of the twenty sites 
i n the field. 

This procedure for collecting and preparing soil samples is the same 
for fields which are smaller than 160 acres, but the number and length 
of transects in the field varies with the acreage. 

All of the nine samples sent to the laboratory from each field 
receive standard agricultural nutrient analysis. This "routine" 
analysis includes measurement of soil pH, electrical conductivity, 
sodium adsorption ratio, organic matter content, and nitrate nitrogen. 
Plant available extract measurements (AB-DTPA) are made for phosphorous, 
potassium, zinc, iron, manganese, copper, lead, cadmium, nickel, 
molybdenum, and chromium. Calcium, magnesium, sodium, potassium, 
chloride, and boron are measured in the saturated extract. 

Three of the samples from each field receive research soil analysis. 
This "detailed" analysis includes total digest measurement (HNO_-HCIO^) 
of arsenic, selenium, copper, zinc, nickel, molybdenum, cadmium, 
chromium, and lead. The total amount of mercury is also measured for 
these samples. 

All of the soils analysis for this project is performed by the 
Colorado State University Soil Testing Lab in Fort Collins, Colorado. 
This laboratory was chosen because of its good reputation in testing for 



29 

selenium and other elements. A report on quality control at the CSU Lab 
is presented in Appendix 2. 

Soil test results for 48 of the sampled fields have been received, 
as of this writing. These results are analyzed from two perspectives 
which are pertinent to the overall project objectives. First, the 
average levels of salt, selenium, and other elements in the soil samples 
are examined. This provides an indication of the variability which 
exists among fields in the water district. This information is useful 
in describing the effects of different drainage scenarios, over time, 
and may be helpful in predicting concentrations of subsurface drainage 
f I ows . 

The variability of salinity levels within a given field is also 
examined. This information will be helpful in developing better models 
of yield responses to saline conditions. Crop fields in the water 
district are large and soil conditions may vary widely within one field. 
Salinity is believed to affect crop yields when a threshold level of 
salts has been exceeded (Maas and Hoffman). Given this framework, it is 
possible that a field with an average soil salinity which is below the 
threshold level may still show a yield effect, if some areas of the 
field exceed the threshold level. 

Soil salinity, sodium adsorption ratio, and boron 

The average levels and the variation of soil salinity and other 
constituents are examined by constructing weighted averages of the eight 
measurements from the top three feet of soil, in each field. The 
weights are determined by the type of sample taken. For example, a soil 
sample from an individual site is given a weight of 20, while a 
composite sample made up of soil from five sites along a transect is 



30 

given a weight of 4. The composite sample which is comprised of soil 
material from all of the 20 individual sites receives a weight of 1. 

Weighted averages and other summary statistics are computed for each 
of the 48 fields. These results are summarized in Table 8. The soil pH 
varies least among the fields examined. The average pH in the district 
is 7.96 and the standard deviation is just 0.17. This results in a 
coefficient of variation of only 2.2. 

The weighted average soil salinity ranges from a low of 1.28 
millimhos per centimeter in one field to a high of 8.45 in another. The 
mean value for the 48 fields is 4.03 millimhos per centimeter and the 
standard deviation is 1.96. The coefficient of variation for the mean 
salinity level is 48.7. The sodium adsorption ratio (SAR) and the 
amount of boron in the soil samples display similar variability. For 
example, the mean SAR varies from a low of 4.00 in one field to a high 
of 17.03 in another. The average of the weighted SAR's is 8.15 and the 
coefficient of variation is 37.6. Average boron levels range between 
0.88 and 5.49 milligrams per liter, with a mean value of 2.29 milligrams 
and a coefficient of variation of 49.1. 

Variation within farm fields is evaluated by examining summary 
statistics for the coefficients of variation which are calculated for 
each field. For example, the coefficients of variation describing soil 
pH range from 0.3 to 3.2, with a mean value of 1.1. This suggests that 
there is little variation in soil pH within fields. The coefficients of 
variation describing soil salinity, SAR, and boron display a much wider 
range of values. Some fields appear to be quite uniform, as indicated 
by the minimum values of 5.7, 4.4, and 11.2. At the same time, other 
fields appear quite variable, as suggested by maximum values of 70.2, 



31 

82.6, and 93.8. The mean values of the coefficients of variation are 

33.7, 30.9, and 37.5 for soil salinity, SAR, and boron. 

Fields which are high in soil salinity tend to be high in boron and 
also tend to have a high sodium adsorption ratio. This is suggested by 
the correlation coefficients calculated for weighted averages of the 
soil samples (Table 9). The correlation coefficients for the means of 
soil salinity, SAR, and boron are all positive and statistically 
significant. Results also suggest that fields which have high levels of 
salt, SAR, and boron tend to have greater variability in these 
measurements. That is, fields with a high average salinity level tend 
to be less uniform than fields with lower average salinity. The same is 
true for SAR and boron. Furthermore, non-uniformity in salinity is 
positively correlated with non-uniformity in SAR and non-uniformity in 
boron . 

Sel en i urn 

The high cost of measuring selenium in soil samples has limited its 
measurement to just three of the nine samples per field which are sent 
to the laboratory. The first sample is a composite made up of some soil 
from each of the 20 sampling sites. The second sample is the one taken 
at the location indicating the highest salinity level in the field, as 
measured with the electromagnetic meter. The third sample which is 
analyzed for selenium is the one taken from the three to six foot soil 
depth. Summary statistics are presented for each of these three 
samples. Results for selenium are compared with those for salinity, 
boron, and the sodium adsorption ratio for these same samples. 

The average level of selenium in the composite soil samples is 1.58 
parts per million (Table 10). The coefficient of variation for selenium 
in these samples is 23.4. The coefficients of variation for salinity. 



32 

SAR, and boron are 43.0, 35.9, and 44.3. This suggests that selenium is 
less variable in these samples than is salinity, SAR, or boron. 

The mean level of selenium measured in samples collected at sites 
indicating high salinity is 1.73 parts per million and the coefficient 
of variation is 35.0 (Table 11). The coefficients of variation for 
salinity, SAR, and boron are 55.7, 48.2, and 55.2 for these samples. 

The average level of selenium in samples taken from the 3-6 foot 
depth is 1.3 parts per million and the coefficient of variation is 27.4 
(Table 12). Again, the variability of selenium in these samples is less 
than that for salinity, SAR, or boron. The mean level of selenium in 
the deep samples is less than that of selenium in the composite samples 
taken from the 0-3 foot depth, as seen by comparing results in Table 5 
with those in Table 10. This is not the case for salinity, SAR, and 
boron. These three measurements have a higher mean value, occur over a 
wider range, and are more variable in the deep samples, than in the 
composite samples from the 0-3 foot depth. These preliminary results 
suggest that while salts and boron may be leached from the upper three 
feet into the lower portion of the soil profile, total selenium may be 
greater in the top three feet than in the 3-6 foot range. 

Correlation coefficients for soil salinity, SAR, boron, and selenium 
are presented in Table 13. Salinity appears to be negatively correlated 



1. Selenium is measured as total selenium in an HNO^-HCIO^ digest, while 
boron and the SAR are measured in the saturation extract. Soil 
salinity is measured in the saturated paste. These differences may 
affect measures of relative variability. 



33 

with soil pH in both the composite and high salinity samples. The 
sodium adsorption ratio and boron are both positively correlated with 
soil salinity in all three types of samples. Total selenium is not 
significantly correlated with salinity in either the composite samples 
or those taken from the 3-6 foot depth. There is a positive and 
significant correlation between soil salinity and total selenium in the 
samples taken from the high salinity locations. 

Electromagnetic Meter Measurements 

Soil conductivity measurements are made at each of the 20 soil 
sampling sites in every field, using a hand-held electromagnetic meter. 
These measurements will be correlated with the soil salinity data 
received from the laboratory to determine a calibration which will allow 
rapid and accurate evaluation of soil salinity in the field. 

The variability of readings obtained with the electromagnetic meter 
is similar to that observed in the laboratory data. For example, the 
average of 20 horizontal mode readings per field ranges from a low of 96 
millimhos per meter to 394 mi I I imhos per meter, with a mean value of 235 
mi II imhos (Table 14). The coefficient of variation is 31.6, suggesting 
moderate variability among fields. Vertical mode measurements display 
simi lar variation . 

The coefficients of variation for the twenty horizontal mode 
readings per field range from a low of 5.9 to a high of 41.7. The 
vertical mode readings are very similar. These results suggest that 
some fields generate very uniform electromagnetic readings, while others 
result in a more variable set of measurements. Uniformity in 
electromagnetic readings is likely correlated with uniformity in soil 
test results from the laboratory. 



34 

Summary 

Soil samples collected from 48 agricultural fields in the Broadview 
Water District during the summer of 1987 have been analyzed for 
nutrients, salts, metals, and selenium. Average soil salinities range 
from a low of 1.3 mi I I imhos per centimeter to a high of 8.4 mi I I imhos. 
Coefficients of variation in individual fields range from 5.7 to 70.2. 
Similar ranges and variability are observed for boron and the sodium 
adsorption ratio. These results suggest that both average levels and 
variability in soil salinity and other constituents may be important in 
constructing farm-level crop production models. 

Fields which are high in soil salinity tend to be high in boron and 
have a high sodium adsorption ratio. High salinity fields are also more 
variable than those which are low in salinity. 

The a-verage selenium level measured in soil samples from zero to 
three feet deep is 1.6 parts per million. The average selenium level in 
samples indicating a high soil salinity is 1.7 parts per million and the 
average level in deep soil samples is 1.3 parts per million. The total 
amount of selenium in soil samples is less variable than soil salinity 
measurements. Total selenium levels are not correlated with soil 
salinity measurements in composite samples or those collected from three 
to six feet deep. 

Measurements of apparent bulk soil conductivity with a hand-held 
electromagnetic meter display variability similar to that observed for 
soil electrical conductivity, as measured in the saturation extract. 
Variability within fields is likely correlated with uniformity in soil 
test results from the laboratory. 



35 

The soils data collected in this study will be used to enhance 
production function analysis and to examine empirical relationships 
between farm-level irrigation strategies and the quantity and quality of 
drain water produced. These and other current efforts in our economic 
analysis of farm-level irrigation and drainage management are discussed 
in the following section. 



36 

5. Current Efforts and Future Work 

The objectives of this research, as described in Section 1, include 
examination of farm-level cultural practices in areas with salinity and 
drainage problems. Production parameters are being estimated from farm- 
level data on soil salinity, irrigation rates, crop yields, and other 
important variables. A farm-level optimization model is being developed 
and will include irrigation and recycling opportunities pertinent to 
typical farming operations in the San Joaquin Valley. 

Several of the research activities described in the first four 
sections of this report represent continuing efforts to develop a data 
base for use in analysis of farm-level drainage management and in 
analyzing the economic impacts of salinity and high water tables on 
growers in drainage problem areas. These efforts, combined with data 
collection and analysis planned for the coming year, will allow more 
accurate prediction and description of farm-level responses to policies 
designed to regulate the quantity or quality of subsurface flows from 
farms or water districts. 
Acti v i ti es 

In particular, future research efforts will include four major 
tasks: 

1. Economic analysis of farm-level decisions regarding optimal 
management of soil salinity and drainage water in the San Joaquin 
Va I I ey . 

2. Estimation of crop-specific production functions for use in a 
regional agricultural production model. 

3. Cooperation with other Drainage Program researchers in examining 
factors which influence the choice of irrigation and drainage 
technologies at the farm level. Specific activities in this task 
will be determined during discussions with cooperating 
researchers. 

4. Relevant data collection activities. 



37 



The farm-level optimization model will consist of three major 
components: 

a. Yield relationships describing crop production as a function 
of applied irrigation water and soil salinity. 

b. Physical relationships describing the movement of salt 
through the soil profile and into drainage and recycling 
systems. 

c. Water quality relationships describing the salt content of 
water at various stages in the system, including surface 
runoff, subsurface drainage, drain water disposed, and drain 
water used in recycling. 

The farm-level objective function specified in the model will be the 
maximization of economic returns to land, management, and fixed 
resources, over time. 

Several drain water disposal scenarios will be examined using the 
optimization model. These will include unconstrained drainage disposal, 
in which case growers are able to choose recycling and disposal rates, 
independent of restrictions on drain water leaving the farm. By 
comparison, quality and quantity constraints on drain water disposal 
will also be examined. Results of this analysis will describe the 
economic impacts which potential constraints may have on the production 
of major crops. Optimal leaching and disposal rates will be determined 
for all scenarios examined. 

Coefficients for use in the regional production model will be 
developed using soils data and other information collected from the 
Broadview Water District. These data will be combined with crop report 
and water delivery data to estimate production functions for major crops 
grown in the district. Estimated functions will be included as key 
physical relationships in the farm-level optimization model. Parameters 



38 

will also be useful in a regional agricultural production model which Is 
based on soils classification data. 

Both the average level of salinity in crop fields and the 
variability of soil salinity will be used as explanatory variables in 
the production functions. 

Data collection efforts in support of the tasks described above, 
will include the following activities: 

a. Collection of water quality data from a set of farm sumps, 
including measurement of electrical conductivity, boron, and 
seleni um. 

b. Collection of field-specific soil salinity data with an 
electromagnetic meter and analyze selected soil samples for 
moisture content, salinity, and other constituents. 

c. Collection of calibration data for use In converting soil 
salinity data, as measured with an electromagnetic meter, to 
estimates of electrical conductivity in the saturated 
extract. 

d. Collection, in conjunction with other Drainage Program 
researchers, of data describing factors which may Influence 
the choice of irrigation and drainage technologies by growers 
in the San Joaquin Valley. 

Discussion 

The soil and water quality data collected during the project will be 
used to enhance estimates of crop-specific production functions. 
Information on average levels of salinity, SAR, and boron will be 
included in the estimated relationships. The variability of these 
measurements within fields will also be included. The amount of 
variation in soil salinity and other elements likely reflects the 
effects of poor drainage and/or non-uniform water applications, over 
time. Inclusion of these measurements In the production function 
analysis will provide more accurate estimates of the relationship 
between crop yields, applied water, and soil quality. 



39 

Additional information on yields and cropping patterns will be 
important in updating and further analyzing the economic impacts of 
salinity and high water tables. Field-specific yield data are being 
collected for crop year 1987. This detail has not been available in the 
past. Yields will be examined for correlations with soil quality and 
cultural practices. 

One of the cultural practices receiving much attention in 
discussions regarding reductions in subsurface flows is pre-i rrigatlon. 
Information obtained during grower interviews has revealed many of the 
economic considerations which determine the amount and timing of water 
applied during winter months. These management concerns need to be 
incorporated in analysis of farm-level responsiveness to changes in 
drainage situations. 

Field-specific data on the amounts and timing of pre-i rrigations 
will allow better description of the farm-level economic considerations 
involved in changing this cultural practice. To date, water use data 
has included only the total amount of water applied to crops. The 
different yield effects of water applied during pre-i rrigations and that 
applied during the growing season can be examined only by collecting 
field-specific data on pre-i rrigations. 

Growers have indicated that some changes in cultural practices 
designed to conserve applied water have already been implemented in 
their farming operations. The motivation behind these changes is the 
economics of crop production. The value of changing the length of 
irrigation runs or irrigating in alternate furrows can be evaluated by 
examining the changes in water use rates and any resulting yield 



40 

effects. Field-specific data on cultural practices, water use, and 
yields will allow this analysis to be performed. 

Field-specific data on the quality and quantity of subsurface flows 
is essential to economic analysis of farm-level responsiveness to 
changes In drainage situations. Changes in cultural practices designed 
to reduce applied water and subsurface flows are not costless. Growers 
will implement these changes only when economic advantages to doing so 
are apparent. Hence, the relationship between changes in applied water 
and the resulting subsurface flows needs to be accurately quantified. 
The relationship between changes in these flows and the resulting 
concentrations and loads of salt and other elements needs to described 
in a way which is pertinent to farm-level operations. The data 
collected in this project include field-level subsurface flows and 
concentrations of salt, selenium, and other constituents. This 
information can be used to accurately predict the farm-level potential 
for changing loads of specific elements. 

Empirical analysis of the relationship between flows and loads 
complements the production function analysis. Yield effects of changes 
in applied water or changes In application strategies can be evaluated 
along with the predicted effects upon quality and quantity of drainage 
water generated. These are precisely the tradeoffs which growers will 
be faced with, in reponse to any changes in drainage situations. A 
complete economic analysis of farm-level drainage management needs to 
include both yield effects of changes in cultural practices and the 
subsurface flow effects (both quality and quantity) of these changes. 
Changes in surface runoff (tall water), as a result of changes in 
farm-level cultural practices also need to be evaluated. Tallwater has 



41 

economic values both in re-circulating for use in crop irrigation and 
in dilution of subsurface drain water, in areas where an outflow quality 
constraint may be placed on agricultural drainage water. The magnitude 
of these two values will be determined by relative prices for irrigation 
water and the costs of treating or reducing subsurface flows. 
Information describing relationships between changes in cultural 
practices, the quality and quantity or subsurface flows, and the 
quantity of tailwater produced will be useful to both farmers and water 
district managers. 
Summary 

Economic analysis of farm-level responsiveness to changes in 
drainage situations requires information on crop yields, cultural 
practices, soil quality, and the quality and quantity of subsurface 
flows and tailwater. When recommendations are made to growers or water 
district managers regarding changes in cultural practices to modify the 
quality or quantity of subsurface drainage water, it will be necessary 
to include estimates of the farm-level economic impacts of these 
changes. This will require describing to farmers, and others, the 
estimated costs of implementing various proposed alternatives and the 
likely farm-level benefits of these changes. These benefits will include 
predicted changes in the quality and quantity of subsurface flows. 
Empirical analysis of the various aspects of farm- 1 eve I drainage 
management described in this report will enhance the accuracy and 
validity of these predictions and the associated recommendations. 



42 



References 



Corwin, D.L. and J.D. Rhoades, "An improved technique for determining 
soil electrical conductivity-depth relations from above-ground 
electromagnetic measurements," Soi 1 Sci . Soc . Am. J . , Volume 46, 
1982. 

Just, R.E., D. Zilberman, and E. Hochman, "Estimation of multicrop 

production functions," American Journal of Agricultural Economics , 
65(4) , November 1983. 

Maas, E. and G. Hoffman, "Crop salt tolerance - current assessment," 

Journal of the Irrigation and Drainage Division , ASCE, 103 (IR2), 
June 1977. 

Rhoades, J.D., "Principles and methods of monitoring soil salinity," in 
Soil Salinity Under Irrigation , Shainberg and J. Shalhevet, eds., 
Spr i nger-Ver lag, 1984. 

Rhoades, J.D. and D.L. Corwin, "Monitoring soil salinity," Journal of 
Soil and Water Conservation, May- June 1984. 



C 

u 

I. 
u 






9 
« 

> 

■D 
.ID 
O 

I. 
CD 



« 
CL 
O 

L « 

U OO 

01 

• 

■P ^ 
U 0) 
« 3 

— O 

• U 

n j: 

I. 

O CM 

f- CO 

o 

• '-I 

I. 

o - 

a -o 

o 

I. ■- 

• I- 
a. c 

a. 

•o « 

• — 

— «J 

>> 

>v 

• J> 

O) 

s > 
■- >> 

• -p 
> c 

• 3 

o 

« 

■P o 

^ c 

0) » 
._ 
« u 



*3 



o 

I. 
9 
O) 
3 
</5 









I. 

a 
m 



a 

<«- -a 
— « 

a c 



« 
o 
4^ 
a 

E 
o 



c 
o 

o 



u. 





s 


U3 


r~ 


s 


'-^ 


CM 


CM 


CM 


CO 


lit 










c 










o 










H 


00 


CM 


U> 


t-» 




IS 


S 


Ul 


s 




CM 


PJ 


CM 


C<J 




<D 


O) 


U> 


CO 


,_^ 


fi 


r^ 


CM 


CM 


« 










c 





















__ 


o> 


CM 


CO 


CM 




f-» 


CM 


CM 


<^ 



CO r~ 00 



T3 

c 

3 
O 

a. 



O) CM CD '« 
rH CM rl CM 



a s S •-< 
^^.^ (0 CO ^ 
c U3 u> r^ 



CO <0 r-l 00 

r* ^ Q fo 
r- ^- CD en 



u. 





s 


T 


00 


1^ 


^ 


CM 


CM 


CM 


CO 


H 










C 










o 










1- 










>M' 


ro 


OO 


ro 


OO 




6> 


CO 


0) 


•w 




CO 


CM 


iH 


o 




CM 


0> 


fH 


U3 


^> 


CM 


r^ 


CM 


CM 



ffi 



▼ CM to 

CM CM CM 



CO 
CM 



a 



o 



« 
a. 



s 


^* 


CM 


OO 


r~ 


t>- 


CD 


CO 


o» 


o> 


o> 


0> 


i-t 


1^ 


ft 


r-t 


1 

CM 


1 


1 
CO 


1 

m 


(0 


r- 


l»- 


00 


O) 


0> 


01 


O) 



n 

« 
« 

L. 

a 
o> 

D 
W 



00 -- » 00 
LO S CO S 

CM <0 00 s 



^ S »< 01 
ID CM r~ to 
CM CO r> s 



c 



4) 



*3 
IS 

o 



u. 



c 
o 

'*> 

u 

3 « 

TJ OO 

O 0> 

I. rH 

o. 

a o> 

o 3 

L O 

U I- 

E -»> 

o 

I. CM 

<•- (O 

0> 

M <-■ 

e 

3 » 

C T3 

9 O 

> — 

• I- 

I. • 

a 

• 



• >> 

• - 

I. c 

« 3 

> o 

< u 



00 
0> 



OO 



CM 






n 



CJ> 



u. 



« 



CD 



— e 
• e 



• 
o 
■p 

a 
E 
o 



ffl 



• 

I. 
u 

< 

a. 

3 



01 



o 



CO 



M 

s 
»- 



c 
o 

o 



I. 
a. 



<0 
OO 



O) 
CM 



CD 



00 S 
iH N 



<0 f^ 
CO OO 
IH CM 



00 

CM 



CM 
(O 



U3 0> 

CM r- 

CO l>- 



rH CM 

U> CO 



•<r 


CM 


^ 


OO 


Ul 


0> 


o 


0) 


0> 


s 


Q 


r^ 



OO S U} CO 

<tj CM CO 1-" 

cv) ^ CO 00 



fO 


CO 


M 


<o 


s 


00 


<0 


00 


T 


* 


r- 


OO 



s 


r~ 


CM 


<0 


r>. 


r^ 


CO 


00 


0> 


01 


0) 


0> 


r-t 


»-t 


ft 


1^ 


1 
CM 


1 


1 
00 


o 


CO 


r- 


t>. 


00 


0> 


(» 


o> 


0» 



45 



Table 3. Increases in average per-acre yields between the third and 
fourth time periods, Broadview Water District and Fresno 
County 









Broadview 


Fresno 


County 


Crop 




Actua 1 


Percent 


Actual 


Percent 


Cotton 




0.3 


Bales 


13.0 


0.4 Bales 


19.0 


Tomatoes 




15.5 


Tons 


80.3 


2.9 Tons 


10.2 


Alfalfa Seed 




337.0 


Pounds 


56.1 


181.0 Pounds 34.2 


Barley 




0.6 


Tons 


33.3 


0.1 Tons 


5.9 


Wheat 




0.6 


Tons 


23.1 


0.3 Tons 


12.0 


Sugarbeets 




4.6 


Tons 


18.0 


2.8 Tons 


10.3 


Note: The 


th 


i rd time period 


i nc 1 udes 


years 1978 thi 


-ough 1982 



and the fourth period includes 1983 through 1986, 



46 



Table 4. Estimated value of yield improvements following installation 
of a drainage outlet in the Broadview Water District 



Estimated Maximum 

Yield Increase due Value of the Yield 

To Improvement in Increase at Average 

Soil and Water Nominal Prices for 

Crop Conditions the period 1983 to 1986 



(Yield per Acre) (Dollars per acre) 

Tomatoes 13.5 Tons 698 

Alfalfa Seed 131.0 Pounds 109 

Barley 0.5 Tons 60 

Wheat 0.3 Tons 37 

Sugarbeets 2.0 Tons 69 



47 



Table 5. Estimated Cobb-Douglas production functions for major crops 
grown in the Broadview Water District, 1962-1986 





Time 


Number 




Constant 


Water^ 


Shift 


Crop 


Period 


of Obs. 


R-Squared 


Term 


Elastic 1 1> 


' F-Statistic 












iPy) 


("2k) 




Barley 


62 


- 70 


84 


.95 


.582 


.120 






71 


- 76 


20 


.89 


.770 


.034 






77 


- 82 


16 


.46 


.264* 


.293 


1.52 




83 


- 86 


6 


.89 


.426* 


.324 


.39 


Cotton 


62 


- 70 


100 


.92 


.725 


.093 






71 


-• 76 


60 


.87 


.714 


.027 


12.05 




77 


- 82 


81 


.84 


.735 


.067 


2.96 




83 


- 86 


43 


.93 


.847 


.086 


3.60 


Sugarbeet 


62 


- 70 


14 


.83 


2.785 


.166 






71 


- 76 


14 


.93 


3.275 


.095 


4.09 




77 


- 82 


5 


.85 


2.686 


.239 


.02 




83 


- 86 


12 


.85 


3.049 


.164 


.38 


Tomatoes 


62 


- 70 


19 


.94 


3.339 


.030 






71 


- 76 


43 


.80 


3.163 


.032 


5.06 




77 


- 82 


19 


.54 


2.845 


.029 


4.18 




83 


- 86 


11 


.74 


3.490 


.033 


12.95 


Alfalfa 


62 


- 70 


71 


.89 


1.850 


.082 




Seed 


71 


- 76 


42 


.75 


1.734 


.104 






77 


- 82 


22 


.56 


1.594 


.067 


3.38 




83 


- 86 


8 


.89 


2.056 


.108 


4.76 


Note: 


The 


superscr i pt 


i ndicates 


that the 


estimated p 


arameter is 

■ 1 



not statistically significant at the five percent level. 

The shift F-statistic pertains to the current and previous 
period models. 



48 



Table 6. Estimated Cobb-Douglas production functions for selected crops 
grown in the Broadview Water District, 1962-1986 



Time Number Constant Water Shift 

Crop Period of Obs. R-Squared Term Elasticity F-Statistic 













Ky^j 


^"2k 


Milo 


62 - 


- 70 


25 


.94 


.419 


.199 


Saf flower 


62 - 


- 70 


12 


.84 


.725 


.093 


Alfalfa 
Hay 


62 - 


- 70 


8 


.81 


1.449 


.140 


Mel ons 


62 ■ 


- 70 


36 


.86 


5.218 


.023 


Dry 
Beans 


77 - 


- 82 


11 


.92 


2.953 


.103 


Wheat 


77 - 
83 - 


- 82 

- 86 


17 
21 


.82 
.95 


.557 
.989 


.283 
.169 



1.88 



Note : The superscript indicates that the estimated parameter is 
not statistically significant at the five percent level. 

The shift F-statistic pertains to the current and previous 
period models. 



49 



Table 7. Actual and estimated water application rates for selected 
crops in the Broadview Water District, 1983-1986 



Actual Average Water Estimated Average Water 
Application Rate Application Rate 

Crop 1985 1986 1983 through 1985 



3.38 (6) 

3.10 (42) 

5.73 (12) 

2.68 (21) 

2.60 (11) 

3.50 (8) 



Barley 


1.91 


(4) 


1.85 


(1) 


Cotton 


3.17 


(28) 


3.06 


(31) 


Sugarbeets 


5.79 


(5) 


4.43 


(3) 


Wheat 


3.28 


(9) 


1.92 


(8) 


Melons 


2.49 


(6) 


2.09 


(5) 


Tomatoes 


3.79 


(5) 


3.10 


(5) 


Alfalfa 










Seed 


2.16 


(4) 


2.66 


(5) 


Alfalfa 










Hay 


3.49 


(2) 


2.56 


(2) 


Corn 


3.75 


(1) 






Dry Beans 






4.18 


(1) 


Others 











3.77 (16) 



Note: The value in parentheses is the number of observations from 
which the average application rate is calculated. 



50 



Table 8. Summary statistics for weighted averages of soil samples 

collected from 20 sites in each field at a zero to three foot 
depth, Broadview Water District, summer 1987 



Number Standard Minimum Maximum Coefficient 
Item (units) of Obs. Mean Deviation Value Value of Variation 



Means 

Soil pH 48 7.96 0.17 7.41 8.29 2.2 

Soi I EC 

(mmhos/cm) 48 4.03 1.96 1.28 8.45 48.7 

SAR 48 8.15 3.07 4.00 17.03 37.6 

Boron 

(mg/l) 48 2.29 1.12 0.88 5.49 49.1 



Coefficients of Variation 



Soil pH 






48 


1.1 


0.7 




0.3 


3.2 


64.8 


Soil EC 






48 


33.7 


16.7 




5.7 


70.2 


49.3 


SAR 






48 


30.9 


16.7 




4.4 


82.6 


53.9 


Boron 






48 


37.5 


16.0 




11.2 


93.8 


42.6 


Notes: 


SAR 


is 


the 


sodium a 


idsorption 


rat 


io. 







Boron is measured as milligrams per liter in the saturation 
extract. 

Selenium is measured as parts per million of total selenium in 

an HNO„-HCIO. digest. 
3 4 



51 



Table 9. Pearson correlation coefficients for weighted averages of soil 
samples collected in the Broadview Water District, summer 1987 



Mean EC Mean SAR Mean Boron CV EC CV SAR CV Boron 



Mean EC 
Mean SAR 
Mean Boron 



CV EC 
CV SAR 
CV Boron 



_„** 


_„** 


.«** 


^«** 


.^ ** 


63 


.83 


.48 


.60 


.41 




.85** 


.51** 


.34** 


.24* 






.56** 


.56** 
.53** 


.45** 

.52** 
.73** 



Notes : SAR is the sodium adsorption ratio. 

The superscript indicates that the correlation coefficient is 
significantly different from zero at the 5-percent level and the 
superscript indicates significance at the 10-percent level. 



52 



Table 10. Summary statistics for composite soil samples collected from 
20 sites in each field at a zero to three foot depth, 
Broadview Water District, one composite sample per field, 
summer 1987 



Number Standard Minimum Maximum Coefficient 
Item (units) of Obs. Mean Deviation Value Value of Variation 



Soil pH 30 7.90 0.20 7.30 8.20 2.5 

Soi I EC 

(mmhos/cm) 30 4.62 1.98 1.50 7.80 43.0 

7.98 2.87 3.50 14.50 35.9 
2.39 1.06 0.90 5.00 44.3 
1.58 0.37 1.00 3.20 23.4 

Notes: SAR is the sodium adsorption ratio. 

Boron is measured as milligrams per liter in the saturation 
extract. 

Selenium is measured as parts per million of total selenium in 
an HNOg-HCIO^ digest. 



SAR 


30 


Boron 




(mg/l) 


30 


Selenium 




(ppm) 


48 



53 



Table 11. Summary statistics for soil samples collected from zero to 

three foot depths at locations indicating high soil electrical 
conductivity, Broadview Water District, one sample per field, 
summer 1987 



Number Standard Minimum Maximum Coefficient 
Item (units) of Obs. Mean Deviation Value Value of Variation 



Soil pH 30 7.97 0.24 7.40 8.40 3.0 

Soil EC 

(mmhos/cm) 30 5.89 3.28 1.40 11.50 55.7 

BAR 30 10.98 5.29 1.70 23.50 48.2 

Boron 

(mg/l) 30 3.36 1.86 0.90 8.10 55.2 

Seleni urn 

(ppm) 48 1.73 0.60 1.05 5.20 35.0 



Notes : SAR is the sodium adsorption ratio. 

Boron is measured as milligrams per liter in the saturation 
extract. 

Selenium is measured as parts per million of total selenium in 
an HNOg-HClO^ digest. 



54 



Table 12. Summary statistics for soil samples collected from three to 
six foot depths in the Broadview Water District, one sample 
per field, summer 1987 



Number Standard Minimum Maximum Coefficient 
Item (units) of Obs. Mean Deviation Value Value of Variation 



Soil pH 44 8.00 0.15 7.50 8.30 1.8 

Soil EC 

(mmhos/cm) 44 5.28 3.33 1.00 14.80 62.9 

SAR 44 10.73 6.34 1.11 25.40 59.1 

Boron 

(mg/l) 44 2.65 1.52 0.50 6.50 57.2 

Seleni urn 

(ppm) 48 1.30 0.36 0.80 2.45 27.4 



Notes : SAR is the sodium adsorption ratio. 

Boron is measured as milligrams per liter in the saturation 
extract . 

Selenium is measured as parts per million of total selenium in 
an HNOg-HClO^ digest. 



55 



Table 13. Pearson correlation coefficients for soil samples collected in 
the Broadview Water District, summer 1987 



Deep 
Samp I es 
(3 to 6 ft) 



-.18 
.87** 
.90** 
.24 



Correlation 
With Soi 1 EC 


Composite 

Samples 
(0 to 3 ft) 


High Sa 1 inity 

Samples 

(0 to 3 ft) 


So i 1 pH 
SAR 
Boron 
Seleni urn 


-.37** 
.55** 
.77** 
.29 


-.36** 
.66** 
.87** 
.36** 



Notes: SAR is the sodium adsorption ratio. 



** 



The superscript indicates that the correlation coefficient is 
significantly different from zero at the 5-percent level. 



56 



Table 14. Summary statistics for the averages of soil electromagnetic 

readings, soil moisture depletions, and soil textures measured 
at 20 sites in each field in the Broadview Water District, 
summer 1987 



Number Standard Minimum Maximum Coefficient 
Item (units) of Obs. Mean Deviation Value Value of Variation 



Means 

Hori zonta I 

(mmhos/m) 48 235.0 74.4 96.0 394.0 31.6 

Vertica I 

(mmhos/m) 48 280.3 85.1 120.5 489.2 30.7 

Soi I Moisture 

Depletion (%) 48 36.2 7.3 17.4 51.6 20.1 

Soi I Texture 

(scale) 48 6.0 1.6 1.0 9.4 26.5 



Coefficients of Variation 



Horizontal 


48 


16.6 


8.1 


5.9 


41.7 




48.7 


Vertical 


48 


17.4 


7.7 


8.1 


41.6 




44.3 


Soi 1 Moisture 


48 


11.5 


6.2 


3.9 


31.6 




53.7 


Soi 1 Texture 


48 


19.3 


10.4 


0.0 


58.8 




53.8 


Notes: Vertical 


and 


horizontal 


measuremen 


its descri 


be the 


app 


arent 



conductivity of the ground, in millimhos per meter. The 
horizontal mode measures conductivity in the top three feet of 
soil, while the vertical mode measures conductivity in the top 
six feet of soi I . 

The soil texture scale assigns a value of 1 to clay, 4 to clay 
loam, 7 to loam, 10 to sand, and 12 to silt. Intermediate 
values are assigned to other classifications on the soil 
triangle. For example, silty clay loam is given a value of 6. 



57 



Appendix 1 

Selected Comments By Growers 

in the 

Broadview Water District 

Provided During Interviews 
August 1987 



58 



GROWER A 

Likes to pre- irrigate in October because the quality of the water is 
better in this month and because there "is no re-circulated subsurface 
water in the system." Tries to push the salts down and keep other 
people from pushing water up on him. So, he pre-i rr igates in October, 
shortly after ripping; "the ground is open and water goes down faster." 

Summer water is saltier. 

On field 1, can't let weed pressure get too great or allow the ground to 
crack — "you'll be short of moisture when ready to plant." 

There has been difficulty with the tile system in field 1. The system 
was dug up two years ago to see what the problem might be, and they 
found that the tile had been laid in the clay rather than in the gravel; 
i.e., it looked like a bad installation. "The system just doesn't put 
out." The operator ties to get as much pre-i rrigation water as possible 
to make allowance for the poor tile system. The strategy is to try to 
push in about 12 inches in the pre-i rrigation, "so that when upslope 
people irrigate, the water won't come down from above." The current 
notion is that with better drainage tile, the pre-i rrigation amount 
could be cut back to about 6 inches. 

Pre-i rrigation is known to push salts up into the beds when prepared for 
cotton. "Some years, when you don't get rain, salts on the top of the 
cotton bed look like snow." 

Last year, on field 1, an effort was made to improve the drainage while 
the leveling operation was going on, by applying three tons of gypsum 
and four tons of chicken manure In rice hulls. They even doubled up on 
the bad spots, but these practices did not seem to make any difference. 

Compliance in the grain program has meant that the rotation is not as 
strong as it might be, "so we will come back with more manure." "I 
don't like to go more than two years In a row on cotton." 

Fertilization strategy is based on a combination of soil tests and field 
history . 

In general, the biggest problem with water is availability (although not 
this year). "We are at the bottom of the district and In past years 
have had delays in deliveries. Often start irrigating a little early to 
give myself a hedge". This problem does not occur In field 2. 

GROWER B 

Doesn't grow wheat because there is no money in it, although it is 
considered to be a clean crop which Is good to grow. 

Summer sugarbeets are irrigated every 2 weeks until one month before 
harvest; in all, over 5 feet of water (on field 3). 

Likes sprinkler irrigation. "There Is no question in my mind, a 
sprinkler program saves money." Thinks that with sprinklers you have 
lower Insecticide costs; also cleaner and less dusty fields, and you can 



59 



get by with one less irrigation. Where land is heavy, furrow irrigation 
drives salts into the seed rows. If you sprinkle, you use less water 
and you have less of a salt problem. We know we have the deep moisture, 
so we sprinkle to establish the stand and then switch to furrow 
i rr igation. 

Salts (high soil salinity) give most trouble on melons and tomatoes, and 
less on cotton and beets. "These crops do real good on salty ground." 

Fertilizer applications are based on soil analysis, although on melons 
sometimes apply fertilizer even if the analysis doesn't call for it. 

Generally the drains are pretty quiet, "but as soon as you put water on 
the field they start running." When not irrigating, more running on 
field 4 than on field 5; however, "run more often on field 6 than on 
field 4." 

It has been observed that the water table is higher on field 4 than on 
field 5. 

"About the only thing that really makes a difference that we can play 
with is the water. This is important!" 

GROWER C 

Fields 7 and 8 are both in the cotton set aside program. 

"When these fields were first purchased, they were level and you could 
grow anything." Since 1975 there have been four or five major floods 
and other nuisance floods. There is lot of silt deposition. On field 8 
the middle is salty, and there are five soil types within a half mile. 
On field 7, the west side blows sand and the east side is adobe clay. 

Attempted to improve soil situation with deep plowing. Ended up 
bringing up the salts and had to leach back down. 

Some years back we grew rice in order to get rid of a morning glory 
problem and to leach down the salts. It worked. 

Next year, will do some leveling and grow alfalfa hay as a reclamation 
crop. A lot of leveling is required, and the strategy is to get the 
flooding figured out before spending very much money on reclamation. 

Sprinkler irrigation is viewed as a way of assuring germination in the 
sand lense areas and helping you through germination if you have salts. 
If you sub-up with lots of salts you end up with the salts right where 
you put the seeds. 

General irrigation strategy as follows: 

(1) If there is not a salt problem, use furrow irrigation. 

(2) If there are salts, then pre-irrigate with sprinklers. Use less 
water but leach the salts out of the seed bed. Pre-irrigating with 
furrow irrigation seems to concentrate the salts while sprinkling 
disperses them. 



60 



(3) If the seed is planted dry, then use sprinklers regardless of the 
sa 1 1 si tuati on . 

"The above discussion refers to cotton. With sugarbeets, never use 
spri nklers. " 

"I know one thing. If we were ever to put a drainage system in again, 
we would have an engineer design it who would be there when the drains 
went in. It would pay to do it!" 

GROWER D 

Went to quarter-mile runs last year. "You can save a little water, 
especially on field 9; you save abut 1/4 of the water." 

Four years ago, because of the water shortage, started watering every 
other row on cotton; liked what it did, and so have adopted it to furrow 
irrigation on cotton. 

The following strategy is used for irrigating cotton: 

(1) 1st irrigation: 12 hours, all rows 

(2) 2nd irrigation: 12 hours, every other row 

(3) 3rd irrigation: 12 hours, every other row 

On the north end of the field 10, there are high concentrations of salt; 
the crop shows more stress, so we give it a 4th irrigation. Also had to 
give an extra irrigation to the sand corner on field 11. 

On field 12, only had to irrigate twice because of the high water table. 

On cotton, "You have to hold the water off and force it to fruit. If 
you keep it too wet (i.e. don't dry it off) it doesn't fruit up." 

On field 13, if it were in cotton, it would not produce very well, you 
would have to irrigate it one more time than field 9. 

If you pull the water, the soil goes to cracking and tears the feeder 
roots. This cuts down production. 

He is of the opinion that you could stretch the length of time between 
irrigations and the yields would not be adversely affected. Thinks that 
you could hold off an extra four or five days after each irrigation, and 
save one irrigation per year. 

After irrigation, we have to wait 14 days to roll back the ditches, 
cultivate and spray. 

"The new varieties of cotton do not do well if they get too wet; they 
will die. " 



61 



General comments on irrigation: 

Would rather pre-irrigate in January. "Then you know that you are going 
to have good, 'live' moisture when you plant. This year, pre-i rrigated 
in early December so we could shut the ranch down." 

On field 9, "When we pre-irrigate, the drainage sump pump can hardly 
keep up." 

Doesn't grow sugarbeets, because they take too much water. 

Since they got rid of the bad (salty) water, things have started to 
change. Cotton was going down hill. 

"Before the salts started to go down, you couldn't make it here on 
tomatoes. You can't make it on 30-32 ton/acre tomatoes. 

Since 1978, you can't make it here in small grains if you include all 
costs! 

Grower E 

Considers himself to be something of an organic gardener. Uses chicken 
manure which he thinks gives him about 40 units of nitrogen per acre, 
along with alot of trace minerals. Also notes that organic farming Is 
not too practical in some respects. 

Says that water is running out of drain on field 14. Last year almost 
100 acre feet. 

In general, notes that the lighter fields on the west side of the farm 
require more irrigation water on the crop; however, it takes more water 
to pre-irrigate the heavier fields on the east side. 

It is his observation that the heavier fields will sub-up better during 
the night because the "moisture comes up better during the night." 

On irrigating wheat: 

Is of the opinion that the smaller the amount of water you use to grow 
the crop, the higher will be the protein content. Thinks that one and a 
half acre feet per acre is plenty of water for a wheat crop. 

The reason he likes a wheat/cotton rotation is because you will get an 
extra 1/4 to 1/2 bale per acre on cotton following wheat, because of the 
value of the wheat crop residue. 

Says that his yields have been coming up since he came here! 

Says that if you work the ground when it is too wet. It will take you 
two years to get it back in shape. 

The strategy for irrigating cotton is as follows: put on 4 to 6 inches 
after it is up, next irrigation 3 to 4 inches, and the third irrigation, 



62 



2 inches. Pre-i rrigation takes about 10 inches. On the average, about 
21 or 22 inches till make a crop. "With careful water management, can 
save about $25 per acre." "In a tight year, that's important." Water 
control is the key. Thinks that the problem of the future will be 
qua I i ty ! 

Can't afford to grow sugarbeets because they take too much water. 

Irrigates cotton in January because "you don't want to get caught short 
and have to come back with irrigation around planting time. Also, after 
it rains, the ground will take more water, so you wait until later in 
the season." 

Question: "Why don't you grow tomatoes?" 
Answer : "I don't want to go broke!" 

Grower F 

With alfalfa seed, you have to keep some of the crop continuously under 
water stress. This induces the plants to flower which, in turn, allows 
the bees to be continuously gathering pollen. 

Pre-i rrigation is not utilized with alfalfa seed. Plant the seed, water 
one time, and establish the stand. You irrigate the whole field the 
first time in late April or early May. Then you wait until some areas 
of the field get stressed before irrigating again. 

Can tell alot of difference in the amount of water in the drains when 
you irrigate. 

His strategy on irrigating cotton is that on the first irrigation, he 
irrigates every other row at about 6" per acre. On the second and third 
irrigations, he irrigates every row at the same 6" per acre rate. 
Thinks that doing every other row facilitates fertilizer uptake. 

Pre-i rrigation is not used on wheat. Plant in furrows, then irrigate. 

Pre-i rrigation is not used on tomatoes. For the last two years, he 
brought tomatoes up on rain water. 

Grower G 

On alfalfa seed: 

With alfalfa seed, it is important to create a water stress in the plant 
in order to cause the plants to flower. However, if there is too much 
stress, you will lose the whole set. It is very difficult to get the 
timing just right. The goal is to try to get three sets of blossoms per 
year . 

In his experience, alfalfa seed does better on the heavier ground (which 
is not so good with cotton). On the lighter soils, it is more difficult 
to control the set. 



63 



The preferred rotation with alfalfa seed is three years of this crop, 
followed by cotton and then wheat. However, because of low wheat 
prices, he is likely to plant cotton for two years in a row. 

He does not use fertilizer on alfalfa seed, figuring that it will become 
established using residual nutrients and then will make its own. 

In general, he feels that he can do well on small grains (over 5 tons 
per acre). However, because of prices, he got into cotton and now is 
restricted on his cotton acreage. The small grains that he grows now 
are grown as part of the rotation, to improve soil conditions. He 
believes that the benefits to the subsequent crop, from the straw 
residue, justify leaving grain in the rotation. 

More on alfalfa seed: 

Historically, there were three alfalfa seed growers in the district. 
They got started to service a seed mill that used to be located in the 
area. The mill has since moved away and now you have to sell your seed 
in other districts. 

There is a lot of variation in alfalfa seed yields due to weather. 
According to growers, if you have to go into October before harvesting 
the crop, "you are in real trouble." "If the rain hits, you could lose 
as much as a third of your crop." Normally, the harvest should start by 
the first of September and be completed by the first of October. 

Tomatoes: 

On field 15, the entire tomato crop was lost, six or seven years ago. 
Went ahead and put in a tile system. Tried tomatoes on this section 
last year and the crop died! Also used to grow tomatoes on field 16, 
but now the salts are too high. On the light soil area of this section, 
the salts are not a problem. However, they are a problem on the heavier 
soil part of the section. 

Crops other than cotton do fine in fields 17 and 18. 

Has use quarter-mile lengths of run for a long time. 

For irrigating cotton, the first irrigation is a 24-hour set. Then use 
12-hour sets for subsequent irrigations. Started doing this about four 
or five years ago in order to cut costs. 

Has not fooled with other types of irrigation because "if you get into 

that, it costs a fortune." However, thinks that sprinklers might be 

good for establishing a stand of alfalfa seed. After the crop is up, 
you would have to use furrows. 

On tile: Where I have tile, I don't see as much salt. 



64 



Grower H 

Has a strategy of applying gypsum at the rate of 3 tons per acre, once 
every three years, over all of his section. Feels he gets better water 
penetration and dissipates the salts. "It mellows out the ground." 
Also has tried sulfur, but found that it did not give any results. 

Next year, he will plant cotton on field 6 and use 30-inch rows, rather 
than the customary 40-inch rows. Feels that the cotton in this quarter 
section is short because of the salts. With shorter spacing, he will be 
growing more plants using the same amount of water, and will thus 
increase yields. 

All of his fields are deep ripped, down to 36", every year. 



65 



Appendix 2 

Quality Assurance Check 

on the 

Colorado State University 

Soi I Testing Lab 

Provided by the California Regional 

Water Quality Control Board 

Central Valley Region 

August 1987 



HUb 1 9 1987 

TE OF CAllfOn'>::A 



GEORGE DEUICMEJIAN. Covcnor 



IIFORNIA REGIONAL WATER QUALITY CONTROL BOARD- 
NTRAL VALLEY REGION 

3 ROUTIER ROAD 
RAMENTO. CA 95S27-3093 




17 August 1937 



Mr. Dan Nelson 
Broadview Water District 
P.O. Box 9 5 
Firebauqh, CA 93 640 



COLORADO STATE UNIVERSITY SELENIUM ANALYSES. 

Enclosed are cooies of selenium analyses conducted by the Colorado 
State University's Soils Testing Laboratory. The samples con 
sisted of duplicate and spiked tile drainage water from the 

The analytical recoveries were 
very good as is outlined in the enclosed copy of a leter sent to 
Steve Workman, the soil's laboratory director. 

From what I understand of your upcoming project, you should be 
manning quality assurance blind duplicates and spiKes to submit 
'to which ever laboratory you choose. If you would 1 i.<e _ .o dis- 
cuss a quality assurance program or need help m obtaining spiked 
water samples, please call me at (916) 361-5689. 



Jeanne Kusick 

Land and Water Use Analyst 

JEK/ jk 



Enclosure 



£ OF CALIFORNIA 



GEORGE DEL KMEJI AN. Go<ferror 



IFORNIA REGIONAL WATER QUALITY CONTROL BOARD- 
TRAL VALLEY REGION 

S STREET 

f^AMEMTO. CALIFORNIA 958".6-7090 
NE: (9i6) 4450270 




24 November 1986 



Mr. Steve Workman 

CSU Soils Testing Laboratory 

Room 6, Vocational Education Building 

Fort Collins, CO 80523 

RESULTS OF SELENIUM WATER S/V-1PLE ANALYSES 

I congratulate your laboratory's selenium analyses on the submitted tile 
drainaae water samples. The recoveries for the samples were very good for .he 
low level spike and still within 10% for the high level spike. Enclosed is a 
copy of the results you sent me with the spike concentrations and expected 
concentrations written in^ 

As you can see, samples JEK850924-2 and -4 were not spiked. These duplicates 
served as the ndicated background concentrations. The recoveries en the 
duplicates ranged from 0.100 mg/1 to 0.108 mg/1 with an average background of 
?03 mg/1. Using 0.103 mg/1 as a basis, expected concentrations were deter- 
mined for ;elenite spikes of 0.250, 0.005, and 0.040 mg/1 (samples -1, -?, and 
-5 respect vely). It was encouraging to note that your procedures were able to 
differentiate a spke concentration as low as 0.005 mc/1. I must however 
qetion the number of replicates analyzed-15 replicates for 5 samples? Uh.t 
is your standard laboratory replicate and spiking procedure? 

You should be receiving five sediment samples within the next waek for parti- 
I?Dat1on n the to" e?^,,,^ ,ound robin project. Good luck with the ana yses 
and pleaslfeel f?ee to call me at (916) 322-6741 if you have any questions. 



JEANNE KUSICK 

Land and Water Use Analyst 

JEK:kwc 
Enclosure 



-y APPROVED 



nor ^U-i.::. " } 





01 












>i 












nj 












^3 


4J 


o 

4-> 






0) o 


o 








^ 


n 


axj 










o 


(0 






iH 


TJ 


u 


rH 






rH 


a> 






• 




•H 


'O 


w 


o 


3 




? 


U: 


•H 


sz 


O 






«J J= 


4J 


>i 




CO 


O 


^ 








0) 


w 




>i^ 




iH 


•H 


C u-i 


c 




OuTS 


o 


•H 


(T3 




e 






4J 


x: 




(0 


>i <u 


O 


El 


\o 


u 




4J 
(0 


c 




00 


Ul 


fl Tt 


3 


• 


CT> 


D 


O 




O 


e 


CN 


o 


•rH 


<D 


>i 


QJ 


\ 


>i 4J x: 




J= O 




(0 


4J 


w 


4-1 


rH 


• « 


e 




w 




• • 


u 


o 


e 


o 


o; 


O 


£-« -u 


o 


1— t 


> 


-U 


O 


D 


u 


c 


i3 


(0 


s 


n} H-> 


3 


n 


Q 



^ 




fO 


«3 




<N 


J 




in 




• 


o 


XJi ITiCO 


C 'U 




•H 


rH 


O rH 


JJ 


CQ 


CJ U3 


CO 


• 


O 


<U ^3 


-IT) 


EH 


U 


to 1 




• 


C r-\ 


iH 


u 


•p-i cr> 


-H 


o 


>H T 


o> 


rH 


w 




O '-^ 




vo CJ en 


D 


t 


• o 


W 


E 


4J n 


U 


(X 


Ct, — - 





U 






o 






(1) 


'a 




<Ti 






4J 


^ 




O 






(0 


ns 




r~ 






•s. 


O 

S3 




1 






r-\ 






rH 






(0 rH 




CO 






c 


O 




in 


VO 




o 


U 




<Ti 


CC 




•H 


4J 






\ 




CP C -P < CN 


^ 


0) 


o 


CJ 


U 


\ 


o « u 


QJ 




o 


•H 






U 


^ 


rH 


w 


nj 


>1-l-> 


o 




3 


•r^ 


4J 


w 


4J 


• • 


\< 


c 


•rH 




c 


u 




u 


rH 


• 


o 


o 


<a 


O 


(0 


w 


e 


K 


C <w 


3 




t3 




c 


•H 


o 


rH 


\A 


0) 


n: rH 




O 


U 


aJ 


(U 


n3 




Cvl 


(13 


O 


>-3 U 




ro 


CO 


Q 



CO 

•H 
>^ 

nj 

C 
< 

(1) 

CJ 

u 
«3 
0) 
W 
0) 



CUV, 
CO cr> 



CO 0) \ 

0) CO CT> 
-u E 

(T3 

u 

•H 
rH 
a rH 

u CO cr> 

e 



CO en 

i. >. 

CT« 

<u CO en 

> e 
to 






o 



o 



O 



I I 

I I 

I I 

I I 

I I : o 



I in fo 
I o in 
I 
I 
o o 



I loom 

1 I rH O ■<* 
I I f-i i-i r-i 
II... 
I I O O O 



o CO CO ro n 

r- O O O 'S' 
en t-{ i—i i-i r-i 

o o o o o 



o o CO o ro 
CTl o o o T 

rt r-i r-{ r-i i-\ 
O O O f3 O 






^ 



O ^ CTi (N VO 

00 o o o rr 

n rH iH rH rH 

o o o o o 



rH CN CO ^ in 



(U 


»* 






rH 


CN 






ao 


<ri 






e M 


o 






nj 








CO 


vo 

00 

:«; 

CiJ 








CN 


f^i 


tT IT) VD 


^ 


CM 


CN 


<N CN CN 


(0 ^ 


CN 


CN 


CN CN CN 


ij 


r-( 


rH 


r— 1 rS rH 







«N e 








1 


• -U O 1 








O rH C 1-1 c 








TJ ^ 


rH 0) IW O 








o 


rH O 








O TD 


^rH > 








CP C VO (DM O 








— (3 


• o to 








03 CO 


rH OJ CO nj 








-P O 










r-i ^ 


- 4J ryv 








3 -H 


r-- o c c 








W Oi 


• C -H (T3 








o to 


o u 








u 


rH to QJ to 








> 


(0 > 4J 








(1) M 


•• ? O rH 








to 


vo CJ (tJ 








o <u 


. to 0) to 








x: CO o^ o u 








4J 


•H O 








^-s 


•• U C rH 








M-l JD 


nH QJ -H J2 








O Oi 


• > 3 








a, 


O O to rH 








^ «— 


rH o e O 








c 


o o to 




^ 




■H i-H 


• • U I-H 




> 




£\ 


to JD c 




1— 1 




4j cn c <u o -H 




Q) 




:5 


O ^ >-i 




u 




3 


•H -rH QjJ^ 




QJ 




O o 


4-) Oi C7> 




O 




>irH 


(3 to to -H 
V-l 3 i^ 




C 

•r4 




4J v;-) 


4J CU O 




CO 




(0 O 


c x: -H <y 








/— 


<p JJ U Ul 








3 IH 


O O nj 








a; 


c y-i t.') 








? XI 


o o to 








o e 


o o o 








C 3 


C C rH 








J^ c 


tji o a. 

C -H rj s 


• 






O (0 


•H to 4-> (3 


ca 






E 


5 -H r3 to 








CD 


o u u 


rH 






^ T3 


f-H 0) -H QJ \ 






(U (C3 


rH U ^O J2 


cn 






•-i e 


O O. C EH 


e 




^ 


o O 


(U 


vo 




0) 


w S 


QJ £ to • 






c 


<a 


^ 4-t 4-) X 


4J 




c 


(U 


4-1 rH -rH 


3 




(a 


rH • 


x: 3 >-l 


O 




0) 


ex. ^ 


^O C7I to 4-) ^ 




fo 


«3 


O 3 QJ ca 
u O u g 


a 




u 


J2 


QJ J= 


c 




n: 




> 4J QJ CJ 


•rH 




o 


u 


O rH ^ j:: 


(0 




Q 


o 


O < 4J 4J 


J-> 











c 

e 
^. 

u 
O 

Q) 
> 

QJ 

4J 

CO 



69 



Figures 



LOCATION MAP OF 
BROADVIEW WATER DISTRICT AREA 




4». Mendota D 



am 



Mendota Q) ^<L ^/^ 



*^^^.^-^ 



O Tranquillity 



\ 



Thousand Acres 





c 




. 


3 




c 


o 







o 




.!-> 






.!-> 


o 




o 


c. 




o 


V) 









CD 


-o 


"0 


O) 


0) 

.4^ 


c 


■^ 


c 


cC 


£ 


CO 


— > 


M 




r> 


J 


IL 


• 1-^ 


o 




-4-) 

r/1 


Lh 

£ 


u 




b- 


< 




CO 


, 




CJ5 


(0 


CO 


'"' 


1— 1 


^ 




0) 


^ 




D 


0) 






> 

CO 
O 

ffl 






oo 

0-) 



oo 

CD 











,:»-^ 








c 






-^ 




^ 


Z3 


r~- 




OJ 


o 


cn 






o 


•» — 




> 








"O 


o 




v_ 


o 


c 




o 


o 


oo 




>- 


03 




CD 




t 


* 


r-^ 












CD 




'' 


^e 


















<r> 


i 













to 

CD 



o CD c:) cr> 

C3 CO CD <Z3 

CZJ CO CD CZ3 

to lO -^ i"0 



c^ 

CO 
CD 



CD CO 

CD CD 

CO CO 



S9J0V 



Thousand Acres 



V) 

o 

CO 



o 

E- 

O 

E- 

-o 
c 



c 

o 
o 

o 

c 

0) 

c 
cd 

o 



en 
0) 

c 
< 



CO 

o 

E- 
I- O 



w 



> 




Csl 
CO 






CD 



o 

CD 

>- 



CD 
CD 







■ -3K 




— ^ 




c: 


s 


Z3 


a> 


o 




CJ> 


> 




-o 


o 


o 


c 


o 


CO 




O) 


CD 


l~ 




Li_ 


< 


> 


-*■ 


< 


> 


J J. 


<P 


■i 








1 



CO 

to 
en 



CO 

en 



S9 JDV 



Thousand Acres 



-a c 

<U o 

^ O 

CD c/3 



CV2 

CO 



-1 



(0 "C 



O 
^ "^ £ 

t- CO 



o 
< 



(D ^ 

M cd 
CD 




CO 



oo 
I — 

CD 









,»N 








— ' 








c: 






5 


Z3 

o 
o 


<yi 




QJ 






> 






k_ 


-o 


o 




D 


o 


£Z 




CJ> 


o 


00 




>- 




<L> 






CO 


La- 


CD 




<> 




1 — 






■5r 


a~) 














Y 


* 











CO 

CO 

en 



CO 

0:1 



S9 JDV 



t: 




c 









) 




u 


CD 


c 


Oi 





'"' 


u^ 


1 




w 


c 




«5 


• ^-4 




-^ 


•o 


o 


o 





a) 


T3 


a 





V 


tn 


F 


a. 


E^ 


>, 


m 


0; 




f»^ 




u 




CO 




CQ 




re 




(Ni 




<u 




^ 




3 




60 






r~ oo 
en en 



I — . oo 
en CT1 



t 9 
<> () 



CZ3 <^Nl 

cnoi 



OnI 1 



■)■ -• 
■i- -• 
■Jr -- 



m 




c 









• »-H 




-I-) 




f ) 


CD 


c 


CO 


D 


■•^ 


Ci. 


1 




w 


c 


o 


o 


05 


• ^^ 




-kJ 


TJ 


u 


O 


D 


u 


T3 







<LI 


in 


£ 


CU 


E^ 


c 




u 




-l_) 




-l-J 




o 




o 




X." 




CM 




0) 




u 




3 




Wj 









C71 



to 

OO 
01 



1 cn 



<> m 
<> () 

^ o 



OO 

cn 



cn cn 



H 



C3 






o 

CD 



m 




C 













u 




C 


<c 





cc 

03 


t- 


'-' 


c 


1 





to 


-kJ 


Oi 


u 


. 


:3 


T3 

n 


T3 


'Z. 


o 


0) 


^ 


D- 


a. 


E 


..-> 




o 


b- 


a; 


cc 



cc 
CO 



CM 



3 




(.O us 
p— CO 
en en 



r~- oo 
CD en 



cn en 



il 



en 

c 
o 



m 




C 




o 




"1 




o 


O 


c 


o 


D 


«— 1 


[ix. 


1 


C 




o 


o 


• f-H 


«-H 


.4-> 


. 






T3 
O 


a. 


u 


E 


CL 




E- 





>v 


.4-J 


a 


CO 













E- 




X 




CN 





a 










C3 



in 






r^ oo 
oi en 



I — oo 
en en 



<> (5 
<> O 



1^ oo 

cncn 



CO r"^ 

cn en 



in 

c 
o 



C 

o 

o 
c 

o 

;^ 



T5 
O 

c 



< 






0) 



60 










o 

o 






to CD 
r^ 00 
en en 



r-« 00 
en oi 



<? <P 



(5 



C3 Cnj 
571 CT> 



cn cr> 



ill 

C5 









* 



t: 




C 









. mm 






CO 


c 


o 







tn 






f^ 


C 









• ^H 


. 


-^ 


■0 


u 


o 


a 




-o 


C- 





c 


s^ 


e 


CL 


E- 


-o 


(Tl 


CD 




0) 




X 




^ 




t4-" 




r-j 




u 




^ 




3 




tl 









r~. CO 
en en 



r^ CO 
C71 en 



t 9 
<> (5 

C ® 



r~ oo 



cn CD 







P00001895